Title: Effect of Two Dissemination Strategies in Improving Private Practitioners’
Adherence to Evidence-based Practice Guidelines for Urinary Tract Infection in a
Developing Country: A Quasi-Experimental Study
Author Name: Mediadora C Saniel
Presenter Name: Mediadora C Saniel
Authors: Saniel MC, Acuin C, Arciaga RS, Lansang MD, Naidas OD, Sevilleja EA, Bustos MV,
Balis AC,Ross-Degnan D
Institution: Philippine Society for Microbiology and Infectious Diseases; International Network
for Rational Use of Drugs
Problem Statement: The effectiveness of strategies for disseminating clinical practice guidelines
(CPGs) and increasing physician compliance with recommended treatment protocols has been
well studied in developed countries but rarely in developing countries.
Objective: Compare the effectiveness of a problem-based lecture-discussion (LD) vs an
interactive case-oriented session (ICS) approach, each combined with feedback on practice data,
as strategies for implementing CPGs on acute cystitis.
Design: Pre/post quasi-experimental design with comparison group.
Setting and Population: Four sites, in the Philippines, each with a cluster of private physicians,
were matched to form two groups. There were 2 clusters of private physicians practising in urban
settings and 2 clusters of private physicians practising in one semi-urban site. The groups were
randomly assigned to either an LD or an ICS intervention. Of the 105 private physicians
participating, 47 were assigned to the ICS group and 58 to the LD group; 98 (93%) completed the
Intervention: Interactive case-oriented sessions with an expert panel were compared with
problem-based lecture-discussions, each followed by feedback and discussions of practice data.
Outcome Measures: Percentage of providers adhering to CPG-recommended antibiotic regimens
for acute cystitis in pregnant and nonpregnant patients. For the multivariate analysis, which also
used hierarchical logistic models, the following were assessed: recommended antibiotic, dosage,
and duration, and laboratory test used according to CPG recommendation. Covariates included
age of physician in years, gender of physician, field of practice (specialization), type of hospital
affiliation (training or non-training), area of practice (urban, semi-urban), number of years since
residency training, and baseline knowledge test score. Multivariate analyses were run twice,
retaining the interval level of quantitative variables in the first run and using the converted
categorical variables in the second to facilitate interpretation of odds ratio.
Results: Appropriate antibiotic regimen prescriptions for acute cystitis in nonpregnant patients
improved from 0% to 3% (odds ratio [OR], 4; 95% confidence interval [CI], 1–18) and to 18%
(OR, 95; 95% CI, 11–817) in LD and ICS groups, respectively, after the educational sessions.
After feedback, adherence further increased to 22% (OR, 57; 95% CI, 16–206) in the LD group
and to 26% (OR, 140; 95% CI, 16–1,205) in the ICS group. For the syndrome of acute cystitis in
pregnant patients, prescribing practices improved significantly only in the ICS group, increasing
from 23% at baseline to 49% after feedback (OR, 2; 95% CI, 1–4). The ORs and the CI’s are
adjusted rates. The confidence intervals for all OR’s range widely because the distribution of
subjects is skewed in the cluster groups.
Conclusions: Private physicians in developing countries can improve their performance if they
are given evidence-based guidelines. Educational strategies that allow interactive discussion and
feedback are more likely to change behavior than one-way communication approaches. Such
strategies should be utilized in continuing medical education programs of specialty societies and
quality improvement efforts in health facilities.
Study Funding: World Health Organization, Management Sciences for Health, INRUD, Pfizer
Philippines Foundation, Inc.